Jobid=626351695560778419 (0.0985)
The Data Management and Biometrics (DMB) group at the University of Twente is seeking one PhD candidate to join the research team of Dr. Nicola Strisciuglio to work on compositional learning for data-efficient vision foundation models.
About the project
Foundation models in computer vision currently rely on massive datasets and brute-force scaling. This leads to high data requirements, hidden biases, limited accessibility, and dependency on corporate-controlled resources. The project aims to develop data-efficient and reliable training strategies for vision foundation models, reducing the need for large datasets and improving robustness.
The project will explore strategies related to 1) Object-attribute compositionality to replace exhaustive data requirements with structured concept learning, 2) Bias detection and machine unlearning to identify and mitigate bias and shortcuts early in training, and 3) Perceptual and conceptual priors to design self-supervised objectives that capture continuous similarity rather than binary contrastive notions.
By embedding compositional structure and prior knowledge into the training process, the project aims to break the dependency on uncontrolled large-scale datasets and enable broader, more transparent development of vision foundation models.
About the PhD position
We offer one fully funded 4-years PhD position, in the context of the project described above, with specific focus on the topic of Compositional Learning for Vision Foundation Models. The aim will be to explore compositional learning strategies (e.g. attribute-object visual representation, or other forms) to reduce the data requirements to train vision foundation models.
When applying, you are required to clearly motivate why you are a good fit for this position and how your expertise and previous experience match the position topic.
Your profile
Our offer
Deel deze vacature: